Telling stories with data – quantitative methods for economical dilemmas

seminaria magisterskie

This seminar’s aim is to help students build their skills necessary for writing their own master thesis by exploring the importance of quantitative methods in understanding and addressing complex economical dilemmas. Students will deepen their understanding of data analytics methodologies and prepare their own studies, which will aim in addressing important research questions in economics. By combining quantitative methods and insightful research ideas, the ultimate goal is to prepare theses which will contribute to the economic literature and popularize the usage of machine learning techniques outside of their traditional context.

In today's data-rich world, an abundance of information is available, providing a wealth of research opportunities to understand and analyze economical phenomena. However, extracting meaningful insights from this vast data pool requires skills in appropriate pre-processing and modeling. This master thesis seminar focuses on data analytics methodologies, with a strong emphasis on machine learning and causal data science techniques. Using data analytics as a powerful tool, students will explore valid research questions relevant to the modern literature in economics. While spatial analysis techniques are encouraged, any topic related to an economical dilemma, such as business activity, consumer behavior, or market analysis, etc. is welcome.

Throughout the seminar, students will be encouraged to focus on drawing insightful conclusions and engaging in discussions about the results, rather than solely applying methods to datasets without thoughtful analysis. The primary objective is to provide meaningful insights into valid research questions in economics, supported by quantitative data analysis methods.

Potential thesis inspirations:

• Evaluation and comparison of different quantitative methods' utility for answering economic questions.

• Case studies on economic data with insights drawn from machine learning techniques.

• Understanding causal relations in economic phenomena using big data sources with causal machine learning.

• Spatial analysis of point (or regional) data and adapting available methods to address spatial data analytic challenges.

During the seminar, students will formulate a valid research question, conduct literature reviews, design their own studies, build narratives using methods and data, and ultimately write their theses. The seminar will consist of individual regular research consultations. The preferred programming language is R.

The desired outcome of the seminar is a publishable paper that contributes to the field of economics and showcases the students' ability to apply quantitative data analysis methods effectively. Through this seminar, students will develop critical skills in leveraging data analytics to tell compelling stories about economical dilemmas and make valuable contributions to the ever-evolving world of economic research.

Selected by advisor, depending on the individual topic of the thesis.

Students can build the quantitative models, analyse the data and draw the conclusions from the conducted research.

Students have a knowledge in R programming and advanced methods of data analysis.

Students can design and develop a project by themselves, are dedicated to work and independent on their research path.

KW01, KW02, KW03, KU01, KU02, KU03, KK01, KK02, KK03

After first (out of 3) semester students have an outline of the thesis prepared, data is collected and hypothesis is prepared.

After second semester (out of 3) literature overview is completed and majority of modelling work done.

After third semester (out of 3) thesis is ready for the defence.